Jan. 20, 1953
EVIDENCE FOR STERIC EFFECT IN
THE
WAGNER REARRANGEMENT
405
(vide s u p a ) , then the electrical effect of the methyl group must also be assumed to be negligible. A negligible electrical effect could be explained on the basis of a resonance interaction between the phenyl and o-tolyl moieties of XIV through the biphenyl (4a-4b) bond. 'The only other plausible explanation for the equal migration ratios of the phenyl Discussion of Results and o-tolyl moieties of XIV would be that the The observed carbon-14 distribution in the prod- electrical and steric effects of the methyl group uct VI1 is good evidence for a steric effect owing almost exactly cancel each other. In the hope to the o-methyl group during the dehydration- that additional data may be gathered which would rearrangement of 2-phenyl-2- (o-tolyl)-ethanol- l-C14 allow a choice between these two explanations, the of 3-methyl-9-fluo(VI). The difference from unity of the ratio (55.0 dehydration-rearrangement is under investigation. I t is f 0.5/45.0 0.5 = 1.22 0.01) of phenyl to renyl-(~arbinol-C'~) o-tolyl migration is clearly much greater than the assumed that the time variable (electromeric) error inherent in the radioactivity determinations and static (inductive) effects of the methyl groups so that the greater per cent. of phenyl migration in this latter compound and in XIV should be nearly identical. is a real and significant value. The rearrangement of carbinols VI and XIV The carbon-14 distribution in the product XV indicates that the over-all influence of the methyl may be formulated as proceeding through resogroup on the dehydration-rearrangement of 1- nance-stabilized intermediates analogous to those methyl-9-fluorenylcarbinol (XIV) is negligible, discussed in the,preceding paper. 1c,21 since the phenyl and o-tolyl moieties in this carbinol Acknowledgment.-The authors are indebted to have migrated in the ratio of 50: 50. This result Drs. G. B. Butler, W. hl..Lauer and J. D. Roberts is quite unexpected in view of the known electrical for many helpful suggestions during the course of effect of a methyl group.2o If the steric effect of this research. the methyl group in XIV is assumed to be negligible (21) This formulation is similar to t h a t supported by Cram to
this material was identified as fluorenone-3-carboxylic acid (lit.19 m.p. 286'). A larger quantity (18 mp,.) of watersoluble acid XXIV was isolated; m.p. 195-200 . This was identified as trimellitic acid. The m.p. of an analytical sample was 206 '. Anal. Calcd. for C~HBOB: C, 51.4; H, 2.86. Found: C, 51.6; H, 2.92; microcuries of carbon-14 per millimole of S X I V , 1.08.
*
*
(19) A. Sieglitz and J. Schalzhes, ibid., 64, 2070 (1921). explain his results in the Wagner-Meerwein rearrangements of iso(20) Drs. J. G. Burr and L. S. Ciereszko in this Laboratory have mers of 3-phenyl-2-butanol and similar compounds: D. J. Cram, THIS shown t h a t the p-tolyl group in 2-phenyl-2-(#-tolyl)-ethanol-l-C*4 JOURNAL, 71, 3863 (1949); D. J. Cram and R. Davis, ibid.,71, 3871 migrates in preference to the phenyl group in the ratio of 66:34 when (1949); D. J. Cram, {bid., 71, 3875 (1949). this carbinol is subjected to the conditions of the Wagner rearrangeOAK RIDGE,TENNESSEE ment.
DIVISIONO F OAK RIDGENATIONAL LABORATORY]
[CONTRIBUTION FROM THE CHEMISTRY
Studies in the Wagner Rearrangement.
V. Further Evidence for a Steric Effect1n2
s. cIERESZK03 AND JOHN G. BURR,JR.
B Y CLAIR J. COLLINS, LEON
RECEIVED JUNE 16, 1952 The radioactivity distributions in the products of the dehydration-rearrangement of 2-phenyl-2-( or-naphthyl)-ethanol-lCI4 and 2-phenyl-2-(~-naphthyl)-ethan0l-l-C'~ have been determined and the per cent. migrations of the phenyl and naphthyl groups have been calculated for each system. These migration ratios are: a-naphthy1:phenyl = 52 :48; a-naphthyl: phenyl = 56:44. The present results are compared with previous work, and are discussed with respect to steric considerations and the presumed carbonium ion intermediates.
Introduction The Wagner rearrangements of 11H-benzo [b]fluorene-1 l-methanol-C14 (I, 52% P-naphthyl migration, 48% phenyl m i g r a t i ~ n ) ~and " of 11Hbenzo [a]fl~orene-ll-methanol-C~~ (11,76% a-naphthy1 migration and 24% phenyl migration)4b (1) This paper is based upon work performed under Contract Number W-7405-eng 26, for the Atomic Energy Commission a t the Oak Ridge National Laboratory. (2) Previous paper, B. M. Benjamin and C. J , Collins, THIS JOURNAL, 7 6 , 402 (1953). (3) Member of the Research Participation Program jointly sponsored by the Oak Ridge Institute of Nuclear Studies and this Laboratory, September, I951-February, 1952. (4) (a) C. J. Collins, J. G. Burr and D. N. Hess, THISJOURNAL, 7 8 , 5176 (1951); (b) C. J. Collins, D. N. Hess, R . H. Mayor, G. M. Toffel and A. R. Jones, ibid., 76, 397 (1953). The migrating groups are connected by a biphenyl-type bond, and thus are not strictly "naphthyl" and "phenyl" groups. I n the interest of ease of expression, however, these fragments, as well as other migrating fragments in the 9-fluorenylcarbinol series, will be referred to by these simpler names.
have been reported. For the latter reaction i t was suggested that the steric influence of the carbon and hydrogen at the l-position of I1 should be either very small or negligible. Next l-methyl-9flu~renyl-(carbinol-C~~) (V) and 2-phenyl-2-(0t ~ l y l ) - e t h a n o l - l - C(VI) ~ ~ were prepared and subjected to the conditions of the Wagner rearrangement.2 While the results indicated a steric effect during the rearrangement of VI, the methyl group in V had no apparent over-all influence on the course of the rearrangement of this carbinol, since the ratio of phenyl to o-tolyl migration was 50:50. This result was unexpected since, if the methyl group of V exerts no steric effect, the electromeric and inductive effects of this methyl group should enhance the migration tendency of the o-tolyl fragment of carbinol V. In the present paper we report the migration ratios during the Wagner rearrangements of
106
CLAIR J. COLLINS, LEONs. CIERESZKO AND JOTIN
c. BURR,JK.
Vol. 7,7
arising from differences in the respective strengths o f the two pinacolic carbon-xygen bonds’“ were circumvented, although such factors as stereospecificityY-“ the variation in degree of migrating group participation in the rate-determining hydroxyl ion rem0va1,~~a and solvent effect13bwere not considered. I I1 I11 Although it would be assumed from Bachmann’s ~ 0 r k 9 a on 1,2-diphenyl-1,2-di-(a-naphthyl)-1,2ethanediol and its /3-naphthyl analog that the “intrinsic migratory aptitudes” of the a- and Pnaphthyl groups are much greater than phenyl, the work of Bergmann and Schuchardtgbshowing stereospecificity in phenyl and a-naphthyl migraIv V tions would cast doubt on these conclusions. Previous attempts14-16 to relate the a-naphthyl and phenyl migratory aptitudes were similarly inconclusive, The migratory aptitude of the o-tolyl group has not been determined since 1,2-diphenyl1,2-di-(o-tolyl)-1,2-ethanediolhas been reported not to react when submitted to the conditions of the pinacol rearrangement. lZc The work presented here, in the paper by Burr and Ciereszko,6 and in three papers by Collins, VI ”I1 VI11 carbinols I11 and IV to the stilbenes VI1 and VIII, et al. , develops an experimental approach whereby respectively. A comparison of these results with the migration ratios of various groups may be determined with a precision of f1%, and without the data for carbinols I and I1 should yield addi- the complications mentioned in a preceding parational evidence for or against a diminished steric graph.” The product yields are essentially quantieffect in the 9-fluorenylcarbinol series, since if the tative during crucial reactions, although the (ja-Gb bond in I1 prevents the carbon and hydrogen determination the of migration ratios is not dependent in the 1-position from exerting a steric effect during upon a quantitative separation of two products from \ r a p e r rearrangement, then the migration ratio a reaction mixture but only upon the carbon-14 of the a-naphthyl group of IV should be signidistribution in a single compound. Thus per cent. ficantly smaller than that of the a-naphthyl moiety of 11. The synthetic and degradative techniques migrations of the a-naphthyl moieties during the were similar to those reported elsewhere for anal- dehydration-rearrangements of carbinols I1 and ogous compounds. ? v 4 3 These results show that IV are 76 and 52, respectively. I t has therefore demonstrated that the presence of the 6a-6b /3-naphthyl migration is preferred over phenyl been (biphenylene) bond in carbinol I1 enhances the migration for carbinol I11 in the ratio 56:44, while ability of the a-naphthyl moiety to migrate. the ratio of a-naphthyl t o phenyl migration for This result is consistent with the suggestion that carbinol IY is 52 :45. the carbon and hydrogen in the 1-position of carbinol I1 exert a very small steric effect during Discussion of Results The rearrangement reactions known as “1,2- the rearrangement of this compound. ‘8 Another shifts”6 have been the subjects of chemical in- possible explanation for these results might be that vestigations for many years. Studies of the during the dehydration-rearrangement of 11, rearrangements of unsymmetrical pinacold pro- intermediate IX, which would be expected to favor duced confusing results chiefly because of the a-naphthyl migration, makes a considerable condifficulties encountered in isolating the principal tribution to the transition st9te. Such an interreaction products8 and in recognizing the existence mediate should be of more importime than X, of stereospecific reacti~ns.~-l During the classical favoring phenyl-group migration. This explanawork with symmetrical pinacols12 the difficulties tion, however, is not considered too likely in view ?e4
( 5 ) J. G. Burr and L. S. Ciereszko, THISJOURNAL, 74, 5426 (1952). (6) See G. W. Wheland, “Advanced Organic Chemistry,” 2d edition, John Wiley and Sons, Inc., New York, N. Y.,1949, pp. 451-534, for an excellent discussion of 1.2-shifts. (7) (a) M. Tiffeneau and J. Levy, Bull. SOL. chim., [IV] 88, 759 (1923); (b) A. M. McKenzie, R. Roger and W. B. McKay, J. Ckem. S o c . , 2597 (1932). (8) A. McKenzie and R. Roger, ibid., 844 (1924). (9) (a) W. E. Bachmann and R. V. Shankland, THISJOURNAL, SI,306 (1929); (b) E. Bergmaun and W. Schuchardt, A m . , 187, 234 (1931). (10) (a) P. I. Pollak and D. Y. Curtin, THISJOURNAL, 72, 961 (19.50); (b) 73, 992 (1951). (11) E. Alexander and D. C. Dittmer, ibid., 73, 1665 (1961). (12) (a) P. J. Montagne, Rcc. trou. ckim., 2 6 , 253 (1907): (b) J. C. Bailar, THISJOURNAL, 62, 3596 (1930); (c) W. E. Bachmann and F. H. hlaser, ; b i d . , 66, 1124 (1932); (d) W. E. Bachniann and J. W. Ferguson, i b i d . , 6 6 , 2081 (1934).
(13) (a) Paper XI, S. Winstein and D. Trifan, i b i d , 74, 1154 (1952), and previous papers; (b) R. F. Brown, ibid.. 74, 428 (1952); R. F. Brown, J. B. Nordmann and M. Mado5, ibid., 74, 432 (1952). (14) A. McKenzie and A. C. Richardson, J . Chcm. Soc., 123, 79 (1923). (15) A. McKende and W. S. Dennler, ibid., 111, 2105 (1924). (16) E. Lucc, Compt. rend., 189, 145 (1925). (17) This technique was apparently first employed t o show that hydrogen migration takes place to the complete exclusion of methyl migration during the pinacol rearrangement of l-methyI-l,2-butanediol; J. D. Roberts, R. E. McMahon and J. S. Hine. THISJOURNAL. 71, 1896 (1949). (18) R. T. Arnold, J. J. Webers and R. M. Dodson, ibid , 74, 368 (1852). have shown that the steric effect of an o-methylene group which is part of a five-membered ring is much less than the steric effect of an emethylene group which is part of a six-membered ring, thus lending support to our suggestion.
Jan. 20, 1953
EVIDENCE FOR STERIC EFFECT IN THE WAGNERREARRANGEMENT
407
1-Phenyl-&( p-naphthyl)-ethylene-l ,2-C114(VII) .-Rearrangement of 800 mg. of carbinol I11 was effected using 2 g. of phosphorus pentoxide and 25 cc. of xylene. The procedure was the same as that previously reported' for the Crystallization preparation of benz[a]anthracene-5,6-C1~~. of the crude product from methanol yielded a first crop of 220 mg. of thick needles, m.p. 146-147' (lit.51 145-146'). A second crop of 150 mg. was obtained from the mother IX x liquors: yield of purified product, 50%. of the observation that &naphthyl migration during Oxidation of O-Phenyl-Z-( 8-naphthyl)-ethylene- 1,2-C1I4 the rearrangement of carbinol I11 occurs to the (VII).-This oxidation was carried out by boiling under reflux for two hours 99 mg. of VII, 1.0 g. of potassium perextent of 56% while &naphthyl migration during manganate, 10 cc. of water and 15 cc. of acetone. This rerearrangement of carbinol I occurs to the extent of action mixture was cooled and passed through a filter. The 52%. Here the net effect of the biphenyl bond in filtrate was concentrated with an air-stream until the acecarbinol I is a repression of the tendency of the 8- tone had been removed, and the remaining solution was chilled and filtered. After one crystallization naphthyl moiety to migrate. The data for the acidified, from methanol the solid naphthoic acid was redissolved in dehydration rearrangements of carbinols I and 111, molar sodium hydroxide, treated with Norit and filtered. in fact, seem to substantiate the case for a large The filtrate was acidified, and the white crystals of 8-naphortho-steric effect in compound IV and a small or thoic acid were filtered from the warm solution; m.p. 180183' with sublimation (lit.** 184-1235'), yield 22 mg. negligible ortho-steric effect in Compound 11. Anal. Microcuries carbon-14 per millimole of a-naphthoic-carboxy-C14 acid, 2.30, 2.26, 2.27. Experimental Phenyl-(~-naphthyl)-acetonitrile-l-C~~.-Thisnitrile was Analytical Determinations.-Carbon-14 determinations prepared in the same manner as the phenyl-(8-naphthyl)were performed by wet combustion to carbon dioxide of 1-5mg. samples of the analytically pure organic compounds acetonitrile-l-C14. Thus from phenyl-(0-naphthyl) -carfollowed by ion-chamber determinations of the carbon-14 binol**10.7 g. of crude phenyl-(a-naphthyl)-~hloromethane2~ content using a vibrating-reed dynamic condenser electrome- was obtained. This was treated with 3.1 g. of cuprous t o yield 5.85 g. (52%) of distilled nitrile. This ter. This method has been described by Neville.lg Stand- cyanide-C14 ardization of ion-current values was made against carbon- distillate crystallized spontaneously. One crystallization C14 dioxide generated from Bureau of Standards sodium from hexane yielded a material whose m.p. was 95-96' carbonate solutions. Melting points are uncorrected. (lit.% 97'). Anal. Microcuries of carbon-14 per millimole, 1.66, 1.66. Carbon and hydrogen assays were performed by Dr. Harry W. Galbraith, Knoxville, Tennessee. 2-Phenyl-Z-(~naphthyl)-acetic-l-C~~ Acid.-The nitrile Phenyl-(p-naphthyl)-a~etonitrile-l-C~~.-Thiscompound was hydrolyzed similarly to the 8-isomer. Thus 2.15 g. of was prepared from phenyl-(8-naphthyl)-carbinol*O and nitrile yielded 2.14 g. (92%) of 2-phenyl-Z-(a-naphthyl)thionyl chloride. The phenyl-(8-naphthyl)-chloromethane a ~ e t i c - 1 - Cacid, ~ ~ m.p. 14O-14l0.P6 was not isolated, but the crude product (80% purity as Anal. Microcuries of carbon-14 per millimole, 1.66. measured by chlorine analysis) prepared from 12 g. of the 2-Phenyl-2-(~-naphthyl)-ethanol-l-C14 (IV).-A lithium carbinol was intimately mixed with 3.0 g. of cuprous cyanide-C". The flask containing this mixture was immersed aluminum hydride reduction of 1.79 g. of 2-phenyl-2-(0in an oil-bath which had been preheated to 200'. During naphthyl)-acetic-l-Cl: acid yielded 800 mg. of purified carbinol IV. m a . 91-92 . the two-hour reaction time the bath was maintained at 200Anal.. Caicd. for C18H160: C, 87.1; H , 6.50. Found: 210'. The cooled melt was extracted with acetone, and after removal of the acetone by evaporation the residue was C, 87.3; H, 6.41. distilled under high vacuum in a short-path still t o give a l-Phenyl-2-(a-naphthyl)-ethylene-l,Z-CI14(VIII).-A prothick yellow oil. This oil was dissolved in acetone and a cedure identical with that used for the preparation of comwhite amorphous residue was removed by filtration. From pound VI1 was employed t o convert 720 mg. of IV to 670 the acetone solution 4 g. (32%) of an oil was obtained which mg. (100%) of the stilbene VIII; m.p. 69-71' (lit.21 m.p. crystallized over a two-day period. One crystallization 70.5' ) . from hexane yielded white prisms, m.p. 85-86'. Oxidation of l-Phenyl-2-(~-naphthyl)-ethylene-l,2-C,14 Anal. Calcd. for ClaHloN: N, 5.76. Found: N. 5.75; (VIII).-This oxidation was carried out analogously to the oxidation of V I I , using 100 mg. of VIII. The yield of purimicrocuries of carbon-14 per millimole, 4.14, 4.18. acid was 26 mg. (35%), m.p. 2-PhenyI-2-(,S-naphthyl)-aceti~-l-C~~ Acid.-The nitrile fied cu-naphth~ic-carboxy-C~~ (3.2 9.) was hydrolyzed in 40 cc. of 50% (by volume) sul- 159-160' (lit.n 159-161'). Anal. Microcurie of carbon-14 per millimole, 0.869, furic acid. The mixture was refluxed for four hours, then cooled, and the oily layer which separated was extracted 0.872, 0.865, 0.875, 0.866 (av. 0.868). with ether. The ether layer was extracted with 2 M soCalculation of Migration Ratios. (a) For the dium hydroxide. Acidification of this alkaline extract of 2-Phenyl-2-(P-naphthyl)yielded a tan oil which subsequently crystallized. This Rearrangement solid was filtered. One crystallization from boiling alcohol ethanol-l-Ci4 (III).-The average millimolar radioafter treatment with Norit followed by a second crystalliza- activity for P-naphthoic-~arboxy-C~~ acid (2.28 pc.) tion from aqueous methanol yielded needles, m.p. 128-130', obtained by the oxidation of VII, when divided by yield of purified material 1.65 g. (50%). Anal. Calcd. for ClaH1402: C, 82.5; H, 5.38. Found: the millimolar radioactivity of I11 (4.10 pc. as C, 82.4; H, 5.49; microcuries of carbon-14 per millimole, determined for 2-phenyl-2-(&naphthyl)-acetic-l-C14 4.09,4.10,4.12, 4.09. acid) gives the factor for P-naphthyl migration. 2-Phenyl-2-(~-naphthyl)-ethanol-l-CLd (III).-This al- Thus, (2.28 + 4.10) X 100 = 55,6% @-iiaphthyl rohol was prepared by the action of 3 g . of lithium aluminum iiiigration (44.4:h phenyl iiiigration). hydride on 1.1 g. of 2-phenyl-2-(j3-naphthyl)-aceti~-l-C~~ acid in ether solution. The excess hydride was decomposed (21) J. L. Everett and G. A . I
